Background: Untargeted metabolomics approaches based on mass spectrometry obtain comprehensive profiles of complex biological samples. However, on average only 10% of the molecules can be annotated. This low annotation rate hampers biochemical interpretation and effective comparison of metabolomics studies. Furthermore, de novo structural characterization of mass spectral data remains a complicated and time-intensive process. Recently, the field of computational metabolomics has gained traction and novel methods have started to enable large-scale and reliable metabolite annotation. Molecular networking and machine learning-based in-silico annotation tools have been shown to greatly assist metabolite characterization in diverse fields such as clinical metabolomics and natural product discovery. Aim of review: We highlight recent advances in computational metabolite annotation workflows with a special focus on their evaluation and comparison with other tools. Whilst the progress is substantial and promising, we also argue that inconsistencies in benchmarking different tools hamper users from selecting the most appropriate and promising method for their research. We summarize benchmarking strategies of the different tools and outline several recommendations for benchmarking and comparing novel tools. Key scientific concepts of review: This review focuses on recent advances in mass spectral library-based and machine learning-supported metabolite annotation workflows. We discuss large-scale library matching and analogue search, the current bloom of mass spectral similarity scores, and how molecular networking has changed the field. In addition, the potentials and challenges of machine learning-supported metabolite annotation workflows are highlighted. Overall, recent developments in computational metabolomics have started to fundamentally change metabolomics workflows, and we expect that as a community we will be able to overcome current method performance ambiguities and annotation bottlenecks.

The evaluation of extraction protocols for untargeted metabolomics approaches is still difficult. We have applied a novel stable isotope-assisted workflow for untargeted LC-HRMS-based plant metabolomics , which allows for the first time every detected feature to be considered for method evaluation. The efficiency and complementarity of commonly used extraction solvents, namely 1 + 3 (v/v) mixtures of water and selected organic solvents (methanol, acetonitrile or methanol/acetonitrile 1 + 1 (v/v)), with and without the addition of 0.1% (v/v) formic acid were compared. Four different wheat organs were sampled, extracted and analysed by LC-HRMS. Data evaluation was performed with the in-house-developed MetExtract II software and R. With all tested solvents a total of 871 metabolites were extracted in ear, 785 in stem, 733 in leaf and 517 in root samples, respectively. Between 48% (stem) and 57% (ear) of the metabolites detected in a particular organ were found with all extraction mixtures, and 127 of 996 metabolites were consistently shared between all extraction agent/organ combinations. In aqueous methanol, acidification with formic acid led to pronounced pH dependency regarding the precision of metabolite abundance and the number of detectable metabolites, whereas extracts of acetonitrile-containing mixtures were less affected. Moreover, methanol and acetonitrile have been found to be complementary with respect to extraction efficiency. Interestingly, the beneficial properties of both solvents can be combined by the use of a water-methanol-acetonitrile mixture for global metabolite extraction instead of aqueous methanol or aqueous acetonitrile alone.

Fusarium graminearum and related species commonly infest grains causing the devastating plant disease Fusarium head blight (FHB) and the formation of trichothecene mycotoxins. The most relevant toxin is deoxynivalenol (DON), which acts as a virulence factor of the pathogen. FHB is difficult to control and resistance to this disease is a polygenic trait, mainly mediated by the quantitative trait loci (QTL) Fhb1 and Qfhs.ifa-5A. In this study we established a targeted GC–MS based metabolomics workflow comprising a standardized experimental setup for growth, treatment and sampling of wheat ears and subsequent GC–MS analysis followed by data processing and evaluation of QC measures using tailored statistical and bioinformatics tools. This workflow was applied to wheat samples of six genotypes with varying levels of Fusarium resistance, treated with either DON or water, and harvested 0, 12, 24, 48 and 96 h after treatment. The results suggest that the primary carbohydrate metabolism and transport, the citric acid cycle and the primary nitrogen metabolism of wheat are clearly affected by DON treatment. Most importantly significantly elevated levels of amino acids and derived amines were observed. In particular, the concentrations of the three aromatic amino acids phenylalanine, tyrosine, and tryptophan increased. No clear QTL specific difference in the response could be observed except a generally faster increase in shikimate pathway intermediates in genotypes containing Fhb1. The overall workflow proved to be feasible and facilitated to obtain a more comprehensive picture on the effect of DON on the central metabolism of wheat.

In this study, a total of nine different biotransformation products of the Fusarium mycotoxin deoxynivalenol (DON) formed in wheat during detoxification of the toxin are characterized by liquid chromatography—high resolution mass spectrometry (LC-HRMS). The detected metabolites suggest that DON is conjugated to endogenous metabolites via two major metabolism routes, namely 1) glucosylation (DON-3-glucoside, DON-di-hexoside, 15-acetyl-DON-3-glucoside, DON-malonylglucoside) and 2) glutathione conjugation (DON-S-glutathione, “DON-2H”-S-glutathione, DON-S-cysteinyl-glycine and DON-S-cysteine). Furthermore, conjugation of DON to a putative sugar alcohol (hexitol) was found. A molar mass balance for the cultivar ‘Remus’ treated with 1 mg DON revealed that under the test conditions approximately 15% of the added DON were transformed into DON-3-glucoside and another 19% were transformed to the remaining eight biotransformation products or irreversibly bound to the plant matrix. Additionally, metabolite abundance was monitored as a function of time for each DON derivative and was established for six DON treated wheat lines (1 mg/ear) differing in resistance quantitative trait loci (QTL) Fhb1 and/or Qfhs.ifa-5A. All cultivars carrying QTL Fhb1 showed similar metabolism kinetics: Formation of DON-Glc was faster, while DON-GSH production was less efficient compared to cultivars which lacked the resistance QTL Fhb1. Moreover, all wheat lines harboring Fhb1 showed significantly elevated D3G/DON abundance ratios.

An untargeted screening strategy for the detection of biotransformation products of xenobiotics using stable isotopic labelling (SIL) and liquid chromatography–high resolution mass spectrometry (LC-HRMS) is reported. The organism of interest is treated with a mixture of labelled and non-labelled precursor and samples are analysed by LC-HRMS. Raw data are processed with the recently developed MetExtract software for the automated extraction of corresponding peak pairs. The SIL-assisted approach is exemplified by the metabolisation of the Fusarium mycotoxin deoxynivalenol (DON) in planta. Flowering ears were inoculated with 100 μg of a 1 + 1 (v/v) mixture of non-labelled and fully labelled DON. Subsequent sample preparation, LC-HRMS measurements and data processing revealed a total of 57 corresponding peak pairs, which originated from ten metabolites. Besides the known DON and DON-3-glucoside, which were confirmed by measurement of authentic standards, eight further DON-biotransformation products were found by the untargeted screening approach. Based on a mass deviation of less than ±5 ppm and MS/MS measurements, one of these products was annotated as DON-glutathione (GSH) conjugate, which is described here for the first time for wheat. Our data further suggest that two DON-GSH-related metabolites, the processing products DON-S-cysteine and DON-S-cysteinyl-glycine and five unknown DON conjugates were formed in planta. Future MS/MS measurements shall reveal the molecular structures of the detected conjugates in more detail.

Fungi can produce a wide range of chemical compounds via secondary metabolism. These compounds are of major interest because of their (potential) application in medicine and biotechnology and as a potential source for new therapeutic agents and drug leads. However, under laboratory conditions, most secondary metabolism genes remain silent. This circumstance is an obstacle for the production of known metabolites and the discovery of new secondary metabolites. In this study, we describe the dual role of the transcription factor Xylanase promoter binding protein 1 (Xpp1) in the regulation of both primary and secondary metabolism of Trichoderma reesei. Xpp1 was previously described as a repressor of xylanases. Here, we provide data from an RNA-sequencing analysis suggesting that Xpp1 is an activator of primary metabolism. This finding is supported by our results from a Biolog assay determining the carbon source assimilation behavior of an xpp1 deletion strain. Furthermore, the role of Xpp1 as a repressor of secondary metabolism is shown by gene expression analyses of polyketide synthases and the determination of the secondary metabolites of xpp1 deletion and overexpression strains using an untargeted metabolomics approach. The deletion of Xpp1 resulted in the enhanced secretion of secondary metabolites in terms of diversity and quantity. Homologs of Xpp1 are found among a broad range of fungi, including the biocontrol agent Trichoderma atroviride, the plant pathogens Fusarium graminearum and Colletotrichum graminicola, the model organism Neurospora crassa, the human pathogen Sporothrix schenckii, and the ergot fungus Claviceps purpurea. Significance Fungi produce a vast number of different chemical compounds via secondary metabolism. These compounds are of great interest because of their potential applicability in medicine, pharmacy, and biotechnology. In addition, a number of such compounds are toxins that potentially represent severe threats to human and animal health. However, under standard cultivation conditions, fungal secondary metabolism remains largely inactive. Here, we show that the deletion of the regulator Xylanase promoter binding protein 1 (Xpp1) results in the production of significantly more secondary metabolites in terms of both number and concentration. Because homologs of Xpp1 exist in fungi with numerous bioactive secondary metabolites, our results can lead to the discovery of secondary metabolites.